Hi Adrian,
On 31/05/16 13:38, Adrian Rodriguez Vilas wrote:
Hello,
Until now I've managed to provide native compilation support. The
flags are passed correctly and the modules generated separately from
the rest.
However, while I was working on this I realized that most of the
scientific software needs workarounds and hacks to be compiled for the
Xeon Phi. The most common (and less problematic) is adding the
"--host=x86-k1om-linux" option when compiling with configure/make,
this one is pretty simple.
In other cases (and not only one or two, I'm afraid) cross-compilation
is not supported by the program, or needs strange hacks that are
incredibly hard to automatize.
In the end, this leaves only a few programs that can be compiled just
with the native support without expending some time with the
particular details of each app. So my question is: is this worth a PR?
It's interesting having this option knowing that most of the
compilations for the Phi will need extra work?
I'm asking this because I don't think I can call it "Xeon Phi support"
knowing that only half the work is done. While the common part of the
Phi builds is included, it will require for developers to create
specific easyconfigs in every application desired.
I think it would still be useful to provide support for building for
Xeon Phi as good as we can on the EasyBuild side.
It may not solve all the problems all the time, but it will alleviate
some of them, and make things easier.
Who knows, it may even help in lettings tools/apps support Xeon Phi
properly...
regards,
Kenneth
2016-05-19 13:21 GMT+02:00 Adrian Rodriguez Vilas
<[email protected] <mailto:[email protected]>>:
My first idea is to get a toolchain compiling for native
execution, as we also have few use for them as accelerators.
Anyway, once I have this done, and if it's interesting, I can try
to make it somehow for offloading mode too.
Regards,
Adrian.
2016-05-17 3:10 GMT+02:00 Christopher Samuel
<[email protected] <mailto:[email protected]>>:
On 16/05/16 22:21, Kenneth Hoste wrote:
> As I already suggested to Adrian, a toolchain option Intel-based
> toolchains to enable building for MIC makes sense.
In offload mode, or as a native binary build?
I ask as we've got some Xeon Phi's but no real applications
that use
them as accelerators - GROMACS had some work started but it's
died now
that Knights Landing is self hosted.
All the best,
Chris
--
Christopher Samuel Senior Systems Administrator
VLSCI - Victorian Life Sciences Computation Initiative
Email: [email protected] <mailto:[email protected]>
Phone: +61 (0)3 903 55545 <tel:%2B61%20%280%293%20903%2055545>
http://www.vlsci.org.au/ http://twitter.com/vlsci
